Node-to-Neighborhood Semantic Consistency: Text-Topology Alignment for TAGs Anomaly Detection

2026-06-29

AI Summary

Existing methods for anomaly detection in text-attributed graphs (TAGs) struggle to jointly model fine-grained textual semantics and graph topology, often overlooking the semantic-topological consistency between a node and its neighborhood. This work formalizes TAG anomaly detection as a node-to-neighborhood semantic consistency problem and introduces the N2NSC framework, which employs a dual-path fusion mechanism to co-optimize large language models and graph neural networks, thereby achieving deep alignment between textual semantics and topological context. Evaluated on eight benchmark datasets, the proposed method significantly outperforms state-of-the-art approaches, effectively identifying anomalous nodes that exhibit inconsistencies in either semantic content or structural connectivity.

Abstract

Graph anomaly detection (GAD) on text-attributed graphs (TAGs) is vital for applications such as fraud detection and academic integrity verification. Existing approaches generally fall into two paradigms. GNN-based methods effectively capture structural patterns but struggle to capture fine-grained textual semantics. Methods integrating LLMs with graphs improve semantic understanding yet fail to fully comprehend topological relationships among neighboring nodes. Moreover, both paradigms overlook the correspondence between textual semantics and graph topological relationships, limiting their ability to identify nodes whose semantics are inconsistent with their neighborhoods. In this paper, we formalize TAG anomaly detection as a node-to-neighborhood semantic consistency problem, where anomalies may arise from either textual semantic mismatch or topological deviation between a node and its neighbors. We propose N2NSC (Node-to-Neighborhood Semantic Consistency), a framework that captures the correspondence between graph topology and textual semantics through two complementary fusion paths. The two pathways work synergistically, enabling the LLM to fully leverage both textual and structural neighborhood information for anomaly detection. Extensive experiments across eight datasets demonstrate that N2NSC consistently outperforms current state-of-the-art methods.

Problem

• text-attributed graphs
• graph anomaly detection
• semantic consistency
• topological relationships
• node-to-neighborhood

Innovation

• Node-to-Neighborhood Semantic Consistency
• Text-Topology Alignment
• Text-Attributed Graphs
• Graph Anomaly Detection
• LLM-Graph Fusion